lIIIL'o°lllll"

ql

.

.2..

DOE/CE/23810-33

MATERIALS COMPATIBILITY AND LUBRICANTS RESEARCHON CFC-REFRIGERANT SUBSTITUTES

' i:?: k

_"&_ _ i _ '_ _ Quarterly MCLR Program _ I] /t 19_4

= :>....='"_o -" _ Technical Progress Report OSTIo_-

_."_,_.___ e. _ 1 October 1993 - 31 December 1993

_"_ _',_._ _,_ _,_

._ = = = = " _', Steven R. Szymurski

t = ,. " _ i .= _ _ Glenn C. Hourahan"-_ ° ,, ° "- _, David S. Godwin8. o_,__ _° ._ >'$

'" _ _ _ r_ _,_

,.., r_ _ _ _g _ _ _ ¢'_

,,-' gh .,.,, ¢2 O __ _ _ _ =._ _ _ Air-Conditioning and RefrigerationqP h Inrv ln©*|hl_',a Inc.=_.- _ _ _ _,e. _, -ec-no.os: ..._...,..,.,

= _ _ ° .. _ o 4301 North Fairfax Drive, Suite 425•. _, _,_, _ ,0 o _ Arlington, Virginia 22203

_Z._ ,_'_ °_ o_

T _, 1994_.=-_ . o _'= _._ January

Prepared for

The U.S. DEPARTMENT OF ENERGY

Office of Building TechnologyGrant Number DE-FG02-91CE23810

This programis supported,in part, by U.S. Departmentof Energygrant numberDE-FGO2-91CE23810:MaterialsCompatibilityand LubricantsResearch(MCLR)onCFC-RefrigerantSubstitutes.Federalfundingsupportingthisprogramconstitutes$6,043,000or 93.67% of allowablecosts.Fundingfrom non-governmentsourcessupportingthis programconsistsof directcost sharingtotaling$408,398or 6.33% ofallowablecosts, andsignificantin-kindcontributionsfrom the air-conditioningandrefrigerationindustry.

Olb, ia!UL_ii_i',_ L)i- i r;_,._ L,,,jGUIVI_NT [d I.JNLIfv'iIFE.D

DISCLAIMER

The U.S. Department of Energy's and the air-conditioning industry's support for theMaterials Compatibility and Lubricants Research (MCLR) program does not constitutean endorsement by the U.S. Department of Energy, nor by the air-conditioning andrefrigeration industry, of the views expressed herein.

NOTICE

This report was prepared on account of work sponsored by the United StatesGovernment. Neither the United States Government, nor the Department of Energy, northe Air-Conditioning and Refrigeration Technology Institute, nor any of their employees,nor any of their contractors, subcontractors, or their employees, makes any warranty,expressed or implied, or assumes any legal liability or responsibility for the accuracy,completeness, or usefulness of any information, apparatus, product or process disclosedor represents that its use would not infringe privately-owned rights.

COPYRIGHT NOTICE

By acceptance of this article the publisher and/or recipient acknowledges the rights of theU.S. Government and the Air-Conditioning and Refrigeration Technology Institute, Inc.(ARTI) to retain a nonexclusive, royalty-free license in and to any copyright coveringthis paper.

TABLE OF CONTENTS

ABSTRACT ......................................................................................... 1

SCOPE ............................................................................................... 1

SIGNIFICANT RESULTS ....................................................................... 2

ON-GOING PROJECTS

Thermophysical Properties of R143a and R152a ............................................. 2Measurement of Viscosity, Density, and Gas Solubility of Refrigerant

Azeotropes and Blends in Selected Refrigerant Lubricants .............................. 7Viscosity, Solubility, and Deasity Measurements of Refrigerant

-Lubricant Mixtures ............................................................................ 8

Sealed Tube Comparisons of the Compatibility of Desiccantswith Refrigerants and Lubricants ............................................................ 11

Accelerated Screening Methods for Predicting Lubricant Performancein Refrigerant Compressors ................................................................... 13

Accelerated Test Methods for Predicting the Life of Motor MaterialsExposed to Refrigerant-Lubricant Mixtures ................................................ 15

Accelerated Screening Methods for Determining Chemical and ThermalStability of Refrigerant-Lubricant Mixtures ................................................ 18

Refrigerant Database ............................................................................. 22

COMPLETED PROJECTS

Thermophysical Properties (R32, R123, R124 and R125) ................................. 24Theoretical Evaluations of R-22 Alternative Fluids ........................................ 30

Chemical and Thermal Stability of Refrigerant-LubricantMixtures with Metals ......................................................................... 32

Miscibility of Lubricants with Refrigerants .................................................. 34Compatibility of Refrigerants and Lubricants

with Motor Materials .......................................................................... 36

Compatibility of Refrigerants and Lubricantswith Elastomers ................................................................................. 39

Compatibility of Refrigerants and Lubricantswith Engineering Plastics ...................................................................... 43

Electrohydrodynamic (EHD) Enhancement of Pool and In-Tube Boilingof Alternative Refrigerants .................................................................... 46

COMPLIANCE WITH AGREEMENT ...................................................... 47

PRINCIPAL INVESTIGATOR'S EFFORT ................................................ 47

DOE/CE/23810-33

MATERIALS COMPATIBILITY AND LUBRICANT RESEARCHON CFC-REFRIGERANT SUBSTITUTES

ABSTRACT

The Materials Compatibility and Lubricants Research (MCLR) program supports critical researchto accelerate the introduction of CFC and HCFC refrigerant substitutes. The MCLR programaddresses refrigerant and lubricant properties and materials compatibility. The primary elementsof the work include data collection and dissemination, materials compatibility testing, andmethods development. The work is guided by an Advisory Committee consisting of technicalexperts from the refrigeration and air-conditioning industry and government agencies. The Air-Conditioning and Refrigeration Technology Institute, Inc., (ARTI) manages and contractsmultiple research projects and a data collection and dissemination effort. Detailed results fromthese projects are reported in technical reports prepared by each subcontractor.

SCOPE

The Materials Compatibility and Lubricant Research (MCLR) program is a multi-year researchgrant administered by the Air-Conditioning and Refrigeration Technology Institute (ARTI), anot-for-profit organization for scientific research in the public interest. The program wasimplemented on 30 September 1991 and, as currently funded, will run through 30 September1995. The MCLR program consists of a number of research projects grouped in phases. Thefirst phase encompasses seven research projects and a data collection and dissemination project.Phase I projects began in January 1992 and were scheduled for completion in March 1993.However, several of these projects have subsequently been extended due to delays or added workwithin the scope of the project. Phase II consists of seven research projects and a data collectionand dissemination project. Phase II projects began in October 1992 and will run throughSeptember 1994. Phase III projects will begin in November 1993 and will run throughSeptember 1995. This report summarizes the research conducted during the 4th quarter ofcalendar year 1993. This report supersedes report numbers DOE/CE/23810-22,DOE/CE/23810-20, DOE/CE/23810-11, DOE/CE/23810-8, DOE/CE/23810-4, DOE/CE/23810-3, DOE/CE/23810-2 and DOE/CE/23810-1.

DOE/CE/23810-33

SIGNIFICANT RESULTS

ON-GOING PROJECTS

THERMOPHYSICAL PROPERTIES OF HFC-143a AND HFC-152a

Objective:

To provide highly accurate, selected thermophysical properties data for refrigerants HFC-143a (CH3CF3) and HFC-152a (CH3CHF2); and to fit these data to simple, theoretically-based equations of state, as well as complex equations of state and detailed transportproperty models.

Results:

The Thermophysics Division of the National Institute of Standards and Technology(NIST) at Boulder, CO, is currently conducting measurements and correlations of HFC-143a and HFC-152a. The new data will fill the gaps in existing data sets and resolve theproblems and uncertainties that exist in and between those data sets. Measurements anddeterminations of thermodynamic properties will include vapor and liquid pressure-volume-temperature (PVT) behavior, saturation and critical points, vapor speed of sound,ideal gas heat capacity, and isochoric heat capacity. The data will then be fitted to theCarnahan-Starling-DeSantis-Morrison (CSDM) and the modified Benedict-Webb-Rubin(MBWR) equations of state. Measurements and correlations of transport properties willinclude thermal conductivity and viscosity. Preliminary results are contained in thequarterly technical progress report, DOE/CE-23810-33A, Thermophysical Properties ofHFC-143a and HFC-152a, 1 October 1993 - 31 December 1993, by W. M. Haynes,PhD. These results are summarized below.

HFC-143a

NIST determined the PVT relationship for the vapor phase of HFC-143a at 121 statepoints using the Burnett apparatus in the isochoric mode. Isochores completed coveredthe following ranges and are depicted in Figure 1:

density: 0.106 to 6.077 mol/L (0.56 to 31.87 lbm/ft 3)pressure: 0.234 to 6.59 MPa (34 to 956 psia)temperature: 276.7 to 373 K (38 to 212°F)

DOE/CE/23810-33

Figure I. Plot of experimental temperatures and pressures for vapor phase PVTmeasurements of HFC-143a [DOE/CE/23810-33A].

7I I I I I

O

6 - • _e

5 - • • -t_ •{2. •

4 - CriticalPolnt_e • -• e • 0 O

,"3- | °°°"-- •

• o• o • gO

2- o|O o°°°

1- $8 • °e°°°e°e°°°°°°•°°°go o• go••••e•••• ° °°-• • O O O • • •

o o, 8 811II II Ill 8 II ;I" • II of" • • of°290 310 330 350 370

Temperature,K

The results of these measurements are tabulated in DOE/CE/23810-33A. NIST

established the densities for the isochores through a Bumett expansion at 373.16 K(212.018°F). Results of the isothermal expansion are also tabulated in DOE/CE/23810-33A.

NIST revised vapor pressure measurements reported in their last quarterly technicalreport (DOE/CE/23810-22A). The results were adjusted for a small air impurity and arepresented in DOE/CE/23810-33A. NIST has reduced the data to the following Wagner-type equation:

In(P/Pc) = TJT(a_r + a_r15 + a3_ 5 + a4r_)

where: r = (1 -T/T0at = -7.34896a2 = 1.70066a3 =-2.02724a4 = -2.65228Tc = 346.23 KPc = 3.7904 MPa

range of validity: 230 to 246.23 K

DOE/CE/23810-33

NIST measured the molar heat capacity at constant volume (Cv) for HFC-143a using anadiabatic calorimeter. Over 250 measurements were conducted on samples in the liquidstate and the liquid and vapor two-phase region. The measurements ranged intemperature from 165 to 343 K (-163 to 158°F) and pressures up to 35 MPa (5100 psia).Preliminary results for the liquid and two-phase region measurements are presented inDOE/CE/23810-33A. Results are graphically depicted in Figures 2 and 3.

Figure 2. Liquid phase isochoric heat capacities of HFC-143a.

x: t

-'(

_ j

50 200 250 300 350

Temperature, K

Figure 3. Saturated liquid heat capacities of HFC-143a.

300

28O

260 o

240 ot

"" 220 o.;..

200 ,)

% 180 ,_,0

- 160 '/

U 140 _joo_,Wj

! 0 0 "</x'z//"X#'c/'d'°°r/# °O:_'_'""

i80 .....................150 200 250 300 350

Tempero[ure, !Z

DOE/CE/23810-33

HFC-152a

NIST has analyzed thermophysical properties measurements from this project and data

from existing literature to develop a 32-term modified Benedict-Webb-Rubin equation ofstate for HFC- 152a.

Table 1. Coefficients to the MBWR Equation of State for HFC-152a (units areK, bar, L, tool).

9 14

p = _ a_p* * exp(-p2/pc 2) E atap2n'lrn-I n-lO

aI = RTa,z = biT + b.zT°'5+ b3 + b4/T + bs/'I_a3 = bsT + b7 + bsT + b9/Tza4 = bloT+ bit+bt2/Tas = b13as = b14 / T + b15/ 'I 2aT = b16/Tas = bl7 / T + bls/T 2as = bl9 / 'Iaalo = b2o / T2 + b,zl / Taall = b22/ T2 + b_ / T4at2 = b,z4/ T2 + b_ / Taat3 = b_s/ T2 + _7 / 'I'4a. = I%81"_+ I_,I'I _als = b3o / 'Ia + bat / 'ra + ba2 / Ts

i b i i b i

1 -0.653136507982x 10"t 17 -0.794945128529x I0"32 0.544321046717x 10 18 -0.1608599400543 -0.121313765605x 103 19 0.617842740721x 10.24 0.181868177534x 106 20 -0.139263903722x 10a5 -0.239916316821x 107 21 -0.533789127839x 10s6 0.131295183149x 10.2 22 -0.906702054774x 1047 -0.314050283774 x 10 23 0.548206634459 x 1098 0.140782204781 x 104 24 -0.938326147547 x 1029 0.292826206065 x 108 25 0.149565454607 x 105

10 -0.244847072578 x 10_ 26 -0.46701114216311 0.70064?7?7825 27:0.627534795462 x 10512 -0.274357889230 x 102 28 -0.943343184078 x 10-213 -0.151880734084 x 10"1 29 0.43215?310903 x 1014 0.206277213827 30 -0.192348324842 x 10415 -0.365615951970 x 10 31 -0.129433862772 x 10-316 0,238408634477 x 10"t 32 -0.233185722528 x 10

DOE/CE/23810-33

NIST reports the equation to be valid at temperatures from 155 to 450 K (-181 to 350°F) and pressures up to 40 MPa (5800 psia). The equation may be reasonablye'_trapolated up to 500 K (440 °F) and pressures up to 100 MPa (14500 psia).

NIST also measured the speed of sound (u) in HFC-152a using a cylindrical acousticresonator. Measurements were conducted along isotherms ranging from 242.8 to 400.0K (-22.7 to 260.3 °F) at pressures from 35 to 1030 kPa (5 to 149.4 psia). Themeasurements are tabulated in DOE/CE/23810-33A. NIST obtained the ideal-gas heatcapacity, Cp°, by analyzing the speed of sound measurements. The results are tabulatedin DOE/CE/23810-33A and were fitted to the following curve:

Cp°/R = ao + a_T + a2T_"+ a3T3

where: SI UNITS

ao = 7.6253 ± 0.0041al (°Cl) -- 0.02021 5:0.00018a2 (°C'2) -" -2.626 x 10s ± 4.6 x 10 .6

a3 (°C3) = 1.035 x 10.7 -t- 2.8 x 10"aR = 8.314471 J/(mol.K)

or PI UNITS

ao = 7.251 _+0.0054al (°Fl) -- 0.01180 ± 0.00014a2 (°F2) = -9.809 X 10 .6 -t- 1.5 X 10.6

a3 (°F 3) = 1.775 x 10-a ± 4.8 x 108R = 0.01419457 Btu/(mol'F)

NIST has begun and is continuing to measure the thermal conductivity of HFC-152a inthe vapor and liquid phases and in the supercritical region. Results should be ready nextquarter.

DOE/CE/23810-33

MEASUREMENT OF VISCOSITY, DENSITY, AND GAS SOLUBILITYOF REFRIGERANT AZEOTROPES AND BLENDS

Objective:

To measure the viscosity, density, and solubility of three refrigerant mixtures that maypotentially replace HCFC-22 or R-502.

Results:

Imagination Resources, Inc., is performing this research under contract with ARTI. Adetailed report of its progress is contained in the quarterly technical progress report,DOE/CE/23810-33C, Measurement of Viscosity, Density, and Gas Solubility ofRefrigerant Blends, 1 October 1993 - 31 December 1993, by Richard C. Cavestri, PhD.

Viscosity, solubility, and density data are reported from -20 to 120 °C (-4 to 248 °F) at69 to 1,724 kPa (10 to 250 psia) for:

• HCFC-22 and Suniso® 3GS mineral oil• R-502 and Suniso ® 3GS mineral oil

• HFC-134a and 32 ISO mixed-acid polyolester• HFC-134a and 32 ISO branched-acid polyolester• HFC-143a and 32 ISO mixed-acid polyolester• HFC-143a and 32 ISO branched-acid polyolester

Also presented in the quarterly report are miscibility screening results of six refrigerantblends with five lubricants. As a result of this screening, the following three refrigerantblends and two polyolester lubricants were selected for evaluation in the balance of theresearch project:

Possible HCFC-22 Alternatives Possible R-502 AlternativeHFC-32/HFC-125 (60/40%) HFC-125/HFC-143a/HFC-134a(44/52/4%)HFC-32/HFC-125/HFC-134a (30/10/60%)

Lubricant..sPentaerythritol ester, branched-acid (ISO 32 cSt)Pentaerythritol ester, mixed-acid (ISO 32 cSt)

It is anticipated that the next report will include complete viscosity, solubility and densitydata on one of the three refrigerant blends.

.................................................................................................................................... ii............ :.......iill/lil ........................................................................................... II....................... I1"....................]11iiI -_

DOE/CE/23810-33

VISCOSITY, SOLUBILITY AND DENSITY MEASUREMENTSOF REFRIGERANT-LUBRICANT MIXTURES

Objective:

To measure the viscosity, solubility, and density of alternative refrigerant-lubricantmixtures

Results:

Spauschus Associates, Inc., is performing this research under contract with ARTI. Adetailed report of its progress is contained in the final report, DOE/CE/23810-34,Solubility, Viscosity and Density of RefrigerantLubricant Mixtures, to be published inMarch 1994, by David R. Henderson, PE.

This research involves viscosity, solubility, and density measurements of thirty-eightrefrigerant-lubricant mixtures listed below at seven different concentrations (0, 10, 20,30, 80, 90, and 100% refrigerant by weight):

Baseline Mixtures:

CFC-12/mineral oil (ISO 32 cSt)CFC-12/mineral oil (ISO 100 cSt)HCFC-22/mineral oil (ISO 32 cSt)

Test Mixtures:

HFC-134a/polypropylene glycol butyl monoether (ISO 68 cSt)HFC-134a/pentaerythritol ester - mixed acid (ISO 22 cSt)HFC-134a/pentaerythritol ester - mixed acid (ISO 32 cSt)HFC-134a/pentaerythritol ester - mixed acid (ISO 68 cSt)HFC-134a/pentaerythritol ester - mixed acid (ISO 100 cSt)HFC-134a/pentaerythritol ester - branched acid (ISO 22 cSt)HFC-134a/pentaerythritol ester - branched acid (ISO 32 cSt)HFC-134a/pentaerythritol ester - branched acid (ISO 68 cSt)HFC-134a/pentaerythritol ester - branched acid (ISO 100 cSt)HCFC-123/mineral oil (ISO 32 cSt)

DOE/CE/23810-33

Test Mixtures (Continued):

HCFC-123/mineral oil (ISO 100 cSt)HCFC-123/alkylbenzene (ISO 32 cSt)HCFC-123/alkylbenzene (ISO 68 cSt)HFC-32/pentaerythritol ester - mixed acid (ISO 22 cSt)HFC-32/pentaerythritol ester - mixed acid (ISO 68 cSt)HFC-32/pentaerythritol ester - branched acid (ISO 32 cSt)HFC-32/pentaerythritol ester - branched acid (ISO 100 cSt)HFC-125/pentaerythritol ester - mixed acid (ISO 22 cSt)HFC-125/pentaerythritol ester - mixed acid (ISO 68 cSt)HFC-125/pentaerythritol ester- branched acid (ISO 32 cSt)HFC-125/pentaerythritol ester - branched acid (ISO 100 cSt)HFC-152a/alkylbenzene (ISO 32 cSt)HFC-152a/alkylbenzene (ISO 68 cSt)HFC-152a/pentaerythritol ester - mixed acid (ISO 22 cSt)HFC-152a/pentaerythritol ester - mixed acid (ISO 68 cSt)HFC-143a/pentaerythritol ester - mixed acid (ISO 22 cSt)HFC-143a/pentaerythritol ester - mixed acid (ISO 68 cSt)HFC-143a/pentaerythritol ester - branched acid (ISO 32 cSt)HFC-143a/pentaerythritol ester - branched acid (ISO 100 cSt)HCFC-124/alkylbenzene (ISO 32 cSt)HCFC-124/alkylbenzene (ISO 68 cSt)HCFC- 142b/alkylbenzene (ISO 32 cSt)

Experimental data for each refrigerant-lubricant mixture in two charts. One presentingthe density as a function of temperature and concentration. The other presentingviscosity and solubility as functions of temperature for given concentrations (Danielchart). Plots will also be generated by fitting the experimental data to mathematicalformulae.

Dynamic viscosity is represented by a modified Walther equation:

log{log(_ + 0.7)} = {al + a2 log(T)} + _{a3 +a4 log(T)} + _o2{a5+ a61og(T)}

where:

/_ dynamic (absolute) viscosity, centipoiseT temperature, Kelvino_ mass fraction refrigerantlog logarithm to the base 10az ... a6 constants

DOE/CE/23810-33

Vapor pressure is represented by:

P = {al + a2T + a3T 2} + ¢o{a4+ asT + a6T2}

where:

P pressure, kilopascalsT temperature, Kelvino_ mass fraction refrigerantal ... a6 constants

Density is represented by:

p = {ai + a2T} + 6o{a3+ aaT} + ¢o2{as+a6T}

where:

p density, gram/cubic centimeterT temperature, Kelvin

, _ mass fraction refrigerantal ... a6 constants

Kinematic viscosity is represented by:

log{log(v + 0.7)} = {at + a21og(T)} + o_{a3+ a41og(T)} + J{a5 + a61og(T)}

where

v kinematic viscosity, centistokeT temperature, Kelvin¢o mass fraction refrigerantlog logarithm to the base 10al ... a6 constants

i

10

.......... i_,.... ....... ....... " 'l¸ ' 1_l ...... _"_........

DOE/CE/23810-33

SEALED TUBE COMPARISONSOF THE COMPATIBILITY OF DESICCANTSWITH REFRIGERANTS AND LUBRICANTS

Objectives:

To provide compatibility information for desiccants with potential substitutes for CFCrefrigerants and suitable lubricants.

To obtain data on chemical and thermal stability of desiccants exposed to refrigerant-lubricant mixtures under anticipated operating conditions.

Results:

Spauschus Associates, Inc., is performing this research under contract with ARTI. Adetailed report of progress in contained in the quarterly technical progress report,DOE/CE/23810-33B, Sealed Tube Comparisons of the Compatibility of Desiccants withRefrigerants and Lubricants, 1 August 1993 - 31 December 1993, by Jay E. Field, PhD.

This project will determine the compatibility of sixteen desiccants in thirteen refrigerant-lubricant mixtures using bench-scale sealed tube tests. Samples will be obtained fromtwo manufacturers for the following eight categories of desiccants:

1. 4A molecular sieve2. 3A molecular sieve3. alumina

4. silica gel5. core type with carbon

10 to 25 % molecular sieve type 3Aalumina5 to 15% carbon

10 to 20% phosphate binder6. core type with carbon

10 to 25% molecular sieve type 4Aalumina5 to 15% carbonI0 to 20% phosphate binder

7. core type without carbon15 to 30% molecular sieve type 3Aalumina

10 to 20% phosphate binder

11

DOE/CE/23810-33

8. core type without carbon15 to 30% molecular sieve type 4AAlumina

10 to 20% phosphate binder

Refrigerant-Lubricant Mixtures Under Study:

1. CFC-11 with naphthen'_c mineral oil2. CFC-12 with naphthenic mineral oil3. HCFC-22 with naphthenic mineral oil4. HCFC-123 with naphthenic mineral oil5. HFC-134a with pentaerythritol mixed-acid polyolester lubricant6. HFC-134a with pentaerythritol branched-acid polyolester lubricant7. HFC-152a with alkylbenzene lubricant (or with pentaerythritol mixed-acid

polyolester lubricant.8. HFC-32 with pentaerythritol mixed-acid polyolester lubricant9. HFC-32 with pentaerythritol branched-acid polyolester lubricant

10. HCFC-124 with alkylbenzene lubricant11. HFC-125 with pentaerythritol mixed-acid polyolester lubricant12. HFC-125 with pentaerythritol branched-acid polyolester lubricant13. HFC-143a with pentaerythritol branched-acid polyolester lubricant

The following tests will be conducted on unexposed desiccant, refrigerant and lubricantsamples and compared with samples exposed for 28 days at 149°C:

Visual InspectionDesiccant Crush StrengthGC Refrigerant DecompositionLubricant Total Acid Number

Liquid Phase Halide Ion/Acid AnionDesiccant Halide Ion/Acid Anion

Preliminary data of on as-received materials and six bead type desiccants in CFC-12 withmineral oil are presented in DOE/CE/23810-33B.

12

• ,, ................ y_ ........ i,_, ........

DOE/CE/23810-33

ACCELERATED SCREENING METHODS FOR PREDICTINGLUBRICANT PERFORMANCE IN REFRIGERANT COMPRESSORS

Objective:

To propose or devise a bench test device for conducting lubricity tests that simulatesconditions in refrigeration and air-conditioning compressors.

Results:

The University of Illinois at Urbana-Champaign is performing this research undercontract with ARTI. An interim report detailing Falex ®comparisons is available underDOE report number DOE/CE/23810-35, Accelerated Screening Methods for PredictingLubricant Performance in Refrigerant Compressors, January 1994, by C. Cusano, PhD.

Refrigerants and lubricz,nts tested in the program are:CFC-12 and mineral oil --- CFC baselineHCFC-22 and mineral c_il --- HCFC baseline

HFC-134a and pentaerythritol ester lubricants --- HFC evaluationR-32/125/134a (30/10/60%) and ester lubricants --- blend evaluation

Comparison of HPT Results with Falex ® Test Results

Qualitative Falex ®results provided by three air-conditioning and refrigeration compressormanufacturers were compared against data measured in the University of Illinois'proprietary high pressure tribometer (HPT). The contact geometries, speeds, andrefrigerant-lubricant mixtures used by the manufacturers in obtaining the Falex ® resultswere modeled in the HPT. However, whereas the Falex ®tests were conducted at room

temperature and atmospheric pressure at relatively high contact loads, the HPT tests wereperformed at temperatures, pressures and load conditions that better approximate criticalcontacts in scroll and reciprocating compressors. The following contact pairs wereevaluated for friction and wear (e.g., wear scars, wear surface, and surface roughness)in unidirectional or oscillating contact tests:

SAE 380 die cast aluminum with carburized 1018 low carbon steel

SAE 356 die cast aluminum with hardened steel drill rod

gray cast iron with SAE 333 die cast aluminum

gray cast iron with carburized 1018 low carbon steel

13

DOE/CE/23810-33

The report draws the following conclusions on the Falex ® and HPT comparisons:

1). For a given refrigerant, varying the speed and the load affects the performanceof a given refrigerant-lubricant combination. Hence, what may be acceptable atone condition may not be acceptable at another operation point.

2). The materials utilized in the contact pairs directly affects the ranking oflubricants. A refrigerant-lubricant combination could have excellent wearcharacteristics with one contact pair and poor wear characteristics with another.

3). In general, for a given contact pair at similar test conditions, the HPT testsranked lubricant performance in an order different than did the Falex ® tests.

4). Generally, no correlation exists between friction and wear. A tribo-contact whichyields relatively low friction can experience relatively high wear, and vice-versa.

Comparison of HPT Results with Compressor Wear Data

To ascertain how well the HPT device models actual compressor operation, it is expectedthat four material contact pairs will be evaluated and the results compared to thoseobserved by compressor manufacturers:

Compressor Application Simulation Possible Contact Pairs

reciprocating wrist pin in conformal 308 die cast aluminum with case hardenedcontact low carbon steel

screw male-female rotor interface in AISI 1141 steel with itself, and with ductilea line contact cast iron

rotary vane-roller line contact sintered ferrous metal with itself

scroll thrust bearing Oldham- aluminum with Norplex TM

coupling area contact

The intent is to determine if the data obtained with the HPT can more accurately predicttribological behavior of critical contacts in compressors than that obtained from simplerFalex ®testers. Utilizing this information, a recommendation will be made on the designof a bench-type device that can be utilized by industry in screening lubricants for usewith various refrigerants.

A report that compares the HPT results against the wear data obtained from compressorlubricity tests will be available in the fourth quarter 1994.

14

DOE/CE/23810-33

ACCELERATED TEST METHODSFOR PREDICTING THE LIFE OF MOTOR MATERIALSEXPOSED TO REFRIGERANT-LUBRICANT MIXTURES

Objectives:

• To develop test methods and procedures to predict the life of motor insulating materialsand varnishes used in hermetic motors.

• To validate proposed test methods and procedures.

Results:

The Radian Corporation has completed Phase 1 of this research under contract withARTI. This phase included a literature search and analysis of current test methods,along with the conceptual design for an improved accelerated test method. Results ofthis study are presented in the report, DOE/CE/23810-21, Accelerated Test Methods forPredicting the Life of Motor Materials Exposed to RefrigerantLubricant Mixtures, Phase1: Conceptual Design, by Peter F. Ellis II and Alan Ferguson, 11 June 1993.

As a result of their studies, researchers at Radian found that the majority of hermeticmotor insulation failures occur in the stator windings of the motor due to a combinationof thermal, chemical, and mechanical interactions. A review of an insurance industrysurvey [Stouppe and Lau, 1989] indicated that 84.0% of hermetic motor failures wereattributed to stator winding failures.

Radian examined several degradation models and investigated the advantages anddisadvantages of the following test methods which are used by industry for testing ofhermetic motors:

Motorette tests (IEEE Standard 117 & UL Standard 984-1989),sealed tube aging tests, andplug-reversal test.

The motorette test uses a simplified simulation of stator windings as the test device. Themotorette is stressed with electrical potential, but no current, while exposed to arefrigerant-lubricant mixture in a heated autoclave. The motorette test method providesinformation on the chemical and thermal degradation of insulation materials. However,it does not provide information of degradation due to the differential thermal expansionor magnetic forces on the windings.

15

DOE/CE/23810-33

The sealed-tube test developed by General Electric [Spauschus and Sellers, 1969;Spauschus and Field, 1979] used bifilar coils of magnet wire sealed in glass tubes withthe refrigerant-lubricant mixtures. Leads of each bifilar coil were sealed through the topof the glass tube, which allowed monitoring of the dielectric properties of the insulation.Although the method was useful for determining the Arrhenius constants of magnet wirevarnish insulation degradation, it does not address the degradation of other insulationcomponents and only simulates the thermochemical aging process.

The plug-reversal test uses a hermetic motor-compressor unit as the test device,modifying the compressor so that it can rotate in either direction with equal ease. Theunit is placed inside a refrigerant loop. The polarities of two of the three phase wiresof the motor are repeatedly reversed, causing the motor to stall and reverse directionwith each reversal. Each plug reversal simulates a locked rotor. This test simulates thefull range of forces on hermetic motors. However, the overall test apparatus is complexand has two drawbacks. Components of the supporting refrigeration test loop often failprior to an actual motor failure and purging the entire test loop for subsequentrefrigerant-lubricant mixture tests is difficult and costly.

A test method has been proposed that combines the advantages of these test methods intoa single practical method. This proposed method uses a stator simulator unit (SSU).The SSU (see Figure 4) consists of a laminated electric steel core, simulating the statorstack of a hermetic motor. The core will contain slot insulation, two coils separated byphase-to-phase insulation and slot wedge insulation.

The test method exposes the SSU to a refrigerant-lubricant mixture in an autoclaveequipped with a headspace chiller and syphon cup similar to those used for motorettetests. Plug-reversal in-rush currents are simulated by intermittent 30 Amp AC pulsesapplied to the lead wires of the SSU.

The SSU and test protocol would emulated the following forces which act on motor statorwindings and cause insulation failure:

thermal agingchemical agingdifferential thermal expansionmagnetodynamic forcestransient voltage stresses from simulated starting cycles.

16

DOE/CE/23810-33

Several parameters will be used to evaluate SSU performance:Winding capacitanceCapacitance (power) dissipation factorSurge testingDC high potential testingPolarization index.

Industry accepted guidelines exist for evaluating each of these parameters which permitdetermination of logical test endpoints, before actually reaching a SSU burnout. It ispostulated that trend analysis results for each of these parameters may allow projectionof the time to a set endpoint well before that end-point is reached. That being the case,then the required test period could be shortened.

The proposed test method will produce results that reflect insulation life relative to areference refrigerant-lubricant mixture. Although Radian concluded that development ofan absolute life prediction test is beyond the state of the art, the proposed SSU testmethod does represent a more economical test method than the battery of methodspresently used by the industry.

Figure 4. Stator Simulator Unit (SSU).

q

............................................................ ,_ ......... __.,._ .... . .......................................... ii1_ ........................... 7.....................................................

DOE/CE/23810-33

ACCELERATED SCREENING METHODSFOR DETERMINING CHEMICAL AND THERMAL STAB_VLITYOF REFRIGERANT-LUBRICANT MIXTURES

Objectives:

• To develop screening methods and procedures to assess the chemical and thermal stabilityof refrigerants and lubricants, as well as additives, metals, surface treatments, andpolymers, used in hermetic systems.

• To validate these screening methods and procedures.

Results:

This research is being performed by the University of Dayton Research Institute undercontract to ARTI.

A literature search has been completed and several analytical techniques that might bedeveloped into accelerated stability screening tests were identified. These methodsemploy one or more of the following techniques:

• Incorporation of thermocouple wells into sample vessels for temperaturemonitoring,

• In situ monitoring of temperature, conductivity, and/or voltage production,• In situ monitoring of viscosity using surface acoustic wavelength devices,• Employing differential thermal analysis (DTA) techniques during sample aging,• Use of flat bottom, four millimeter diameter glass tubes for sample analysis,• Use of miniature metal bombs for sample analysis.

The report, DOE/CE/23810-10, Accelerated Screening Methods for DeterminingChemical and Thermal Stability of Refrigerant-Lubricant Mixtures; Part I: MethodAssessment, by Robert Kauffman, April 1993, gives more details on the results of thisliterature search and the candidate screening methods. This report is currently availablefrom the ARTI Refrigerant Database (RDB3501, 42 pages).

Part II concentrates on evaluating various techniques for development into an acceleratedscreening method. Details of the contractor's progress are contained in the quarterlytechnical progress reports, DOE/CE/23810-20D (1 March 1993- 30 June 1993),DOE/CE/23810-22D (1 July 1993- 30 September 1993), and DOE/CE/23810-33D(1 October 1993 - 31 December 1993), Accelerated Screening Methods for Determining

18

DOE/CE/23810-33

Chemical and Thermal Stability of Refrigerant-Lubricant Mixtures; Part II: ExperimentalComparison and Verification of Methods, by Mr. Robert Kauffman. The latter reportis still undergoing technical review and will be available when approved.

Tests employing DTA techniques, using thermocouples or thermistors inside or outsidethe sample vessels, have been conducted. Initial results indicate that these techniques areonly slightly sensitive to CFC-12/mineral oil reactions. It is hypothesized that thesetechniques will be less sensitive to HCFC/lubricant and HFC/lubricant reactions.

Use of ferric fluoride as a degradation catalyst was tested. Initial results show that attemperatures above 175°C (347°F), the catalyzed reactions appear to be more dependenton lubricant degradation than on refrigerant degradation. It is concluded that the use offerric fluoride as a catalyst may have the potential for development into an acceleratedscreening method for lubricant stability.

In situ color (light transmission) measurements were tested as a potential stabilityscreening method. It was found that transmission depended on temperature and lightsource output, as well as color change of the refrigerant-lubricant mixture, and thereforemay not be as promising as other screening techniques reviewed.

Tests involving in situ conductivity monitoring have also been performed. Thesetechniques involve measuring current between two metal electrodes, sealed into thesample vessel, with a known applied voltage. Evaluations were made using combinationsof: ac or dc voltage; tungsten, copper, and/or iron metal electrodes; steel, copper or nometal coupons as catalysts; and continuous or non-continuous conductivity monitoring.Initial results indicate that the in situ conductivity measurements correlate withrefrigerant-lubricant stability as reported in the literature and as determined by otheranalytical techniques (color and gas chromatography measurements). Initial results alsoshow that continuous measurement of conductivity (i.e., maintaining the applied voltagethroughout the aging process) accelerates as well as monitors the degradation ofrefrigerant-lubricant mixtures.

Tests were conducted using HFC-134a and four polyolester lubricants, heated in modifiedglass tubes (see Figure 5 below) for two days at 175°C (347°F). Conductivity wasmonitored continuously by application of a triangular voltage wave-form across twotungsten leads sealed into the tubes. Dramatic changes in the first twelve hours ofmeasurements are hypothesized to be related to interactions between the metal (tungsten)surface and the refrigerant-lubricant mixture. Conductivity changes thereafter (between12 and 48 hours) were seen to correspond to chemical/thermal stability as determined byASTM color tests:

19

....................................................................... _-'_'_ .................................................. .]........ I1:I1'11..........I inJ .........

DOE/CE/23810-33

Table 2. Analytical Results for Aged HFC-134a/Lubricant Mixtures

ASTM Color Conductivity Percent(absolute units) Change

I

Lubricant Before After T = 12 hours I T=48 hours absoluteI I Illl I IllIlll Illl _

Pentaerythritol <0.5 <0.5 828.36 921.31 11.2Branched-Acid 2

Pentaerythritol <0.5 0.5 186.67 116.36 37.7Mixed-Acid 2

Pentaerythritol < 0.5 < 1.0 438 268 38.8Branched-Acid 1, "fresh"

Pentaerythritol 0.5 3.5 707 1033 46.1Branched-Acid 1, "exposed"

Figure 5. Modified Sealed Glass Tube.

-_ _.. Refrigerant/LubncantMixture

Metal Wire _.. /Electrodes

"- _Glana20 to TungstenSeals

DOE/CE/23810-33

Two aluminum heating blocks have been constructed with built-in cartridge heaters andelectrical connections for monitoring the conductivity of the fluids inside the modifiedsealed tubes. A programmable temperature controller will be used to subjectrefrigerant/lubricant mixtures to both isothermal and ramped temperature tests. Figure 6,below, is a schematic of a three-well aluminum block heating system.

Figure 6. Three-Well Aluminum Block Heating System.

/ Screws to

PressureRelier ... PressureInlet /S_ure Lid

_Prmuriz_l_._,__ _,d

Pressure Tight Solid Aluminum orSeal mphile Bhxk

IleatingCartridge

_9

for Co_llucllvt_yMeuufemeflui

Electrical Connections for

Conductivity Measurements

21

DOE/CE/23810-33

REFRIGERANT DATABASE

Objectives:

• To develop a database for materials compatibility and lubricant research (MCLR)information on substitutes for chlorofluorocarbon (CFC) and hydrochlorofluorocarbon(HCFC) refrigerants for applied refrigeration cycles.

• To assemble physical properties, materials compatibility, and related test data for theserefrigerants and lubricants, along with comparative data for currently-used refrigerants.

• To make the data readily accessible for rapid screening and identification of pertinentsource documents based on user-defined search criteria.

Results:

James M. Calm, Engineering Consultant, is performing this research under contract toARTI. The database is available on a subscription basis (for a nominal charge to recoverdistribution costs) in either a computerized or printed format.

The core of the database consists of bibliographic citations and extended abstracts forpublications that may be useful in research and design of air-conditioning andrefrigeration equipment. The bibliographic citations provide information to facilitateordering of source documents from the author or the publisher. Approximately 40% ofthe documents are available from the database contractor. Detailed abstracts have been

prepared for many of the entries. These detailed abstracts describe the data, tests,evaluations, and the materials noted in the documents. The abstracts permit searchingof information by refrigerant or refrigerant-lubricant combination, topic, author, material(by generic or commercial name), specific refrigerant property, or just about any othercombination of search criteria.

The computerized version of the database includes summaries for over 175 refrigerants,both single-component and blends. Refrigerants are identified by ASHRAE Standard 34designations, chemical names and formulae, common names, refrigerant groups, blendcompositions, and familiar chemical abstract numbers. Summary property data (withdimensional quantities in dual IP and SI units) are provided for molecular mass,atmospheric boiling point, melting or freezing point, and critical-point parameters(temperature, pressure, specific volume, and density). The lower an6 upper flammabilitylimits (LFL and UFL), ASHRAE Standard 34 safety classification, ozone depletionpotential (ODP), global warming potential (GWP), halocarbon global warming potential

22

|

DOE/CE/23810-33

(HGWP), and common uses are indicated if known. Specific sources are referenced forthe data to enable verification, obtaining further information, and examining underlyinglimitations.

The February 1994 release of the ARTI Refrigerant Database will contain in excess of1,350 entries related to:

• refrigerant properties• performance with new refrigerants• materials compatibility• lubricants for new refrigerants• environmental and safety data• related research programs

23

DOE/CE/23810-33

COMPLETED PROJECTS

THERMOPHYSICAL PROPERTIES (HFC-32, HCFC-123, HCFC-124 AND HFC-125)

Objective:

To provide highly accurate, selected thermophysical properties data for refrigerants HFC-32, HCFC-123, HCFC-124, and HFC-125; and to fit these data to simple, theoretically-based equations of state, as well as complex equations of state and detailed transportproperty models.

Results:

The Thermophysics Division of the National Institute of Standards and Technology(NIST) has completed measurements and correlations of HFC-32, HCFC- 123, HCFC- 124and HFC-125. This data filled the gaps that existed in data sets and resolved problemsand uncertainties that existed in and between those data sets. Measurements and

determinations of thermodynamic properties included vapor pressure-volume-temperaturebehavior, liquid pressure-volume-temperature behavior, saturation and critical points,vapor speed of sound and ideal gas heat capacity, and isochoric heat capacity. The datawas fitted to the Carnahan-Starling-DeSantis (CSP) and the modified Benedict-Webb-Rubin (MBWR) equations of state. Measurements and correlations of transportproperties included thermal conductivity and viscosity measurements.

A detailed report of the results is presented in the final report, DOE/CE/23810-16,Thermophysical Properties, April 1993, by Richard F. Kayser, PhD (RDB #3860, 242pages). Key results are summarized below:

HFC-32

NIST has developed a 32-term MBWR equation of state (Table 1) for HFC-32. Theequation is reported to be valid at temperatures from the triple point at 137 K (-213°F)up to 400 K (2600F), and it may be reasonably extrapolated up to 500 K (440°F). Themaximum pressure for the equation is 40 MPa (5800 psi), and it may be reasonablyextrapolated up to 100 MPa (14500 psi). NIST fitted the equation using a multi-parameter linear least squares routine on the measured data.

24

DOE/CE/23810-33

HCFC-123

NIST has revised the MBWR equation of state for HCFC-123. This work was promptedby an evaluation of the equations of state for HFC-134a and HCFC-123 carried out byAnnex 10 of the International Energy Agency. Weaknesses revealed during theevaluation included the derived properties for speed of sound and heat capacity. Therevised equation (Table 2) is reported to be valid at temperatures from just above thetriple point up to 550 K (530°F) and at pressures up to 40 MPa (5800 psi).

HCFC-124

NIST has developed a 3_-term MBWR equation of state (Table 3) for HCFC-124. Theequation is reported to be valid at temperatures ranging from 210 to 450 K (-82 to350°F) and it may be reasonably extrapolated up to 500 K (440°F). The maximumpressure for the equation is 20 MPa (3000 psi).

HFC-125

NIST has developed a 32-term MBWR equation of state (Table 4) for HFC-125. Theequation is reported to be valid at temperatures ranging from 200 to 400 K (-100 to260°F). It may be reasonably extrapolated up to 500 K (440°F). The maximumpressure for the equation is 20 MPa (2900 psi).

25

DOE/CE/23810-33

Table 3. Coefficients to the MBWR equation of state for HFC-32(units are K. bar. L0 moi)

9 15

p-En=l n-10

Oc = 8.1245 mol/L

a1 = RTa2 = biT + b2To.5 + b3 + b4/T + bs/T2a3 = b6T + b7 + bs/T + bg/T2a4 = bto T + bll + bt2tTa5 = bI3 _

a6 = bt4/T + bls/T 2a7 = bt6/Ta8 = bt7/T + bts/T 2at) = bl9/T 2alo = b20/T2 + b21/T3all =: b22/T2 + b23/T4a12 = b24/T2 + b25/T3a13 = b26/T2+ b27/T4a14 = b28/T2 + b29/T3al5 = b30/T2 + b31/T3 + b32/T'*

i b i

I -0.184799147712E-01 17 -0.399464119357E-042 O.199258716261E+01 18 0.653548292730E-013 -0.45081,8142855E+02 19 -0.119312200130E-024 0.517320130169E+04 20 -0.896057555372E +055 -0.770847082500E +06 2I -0.218872108921E +086 -0.170184611963E-03 22 -0.189705435851E+047 -0.143023459131E+01 23 0.310718784685E+088 0.606314008455E +03 24 -0.126638710844E +029 0.192559574847E+06 25 0.246519270465E +04

10 -0.596044051707E-04 26 -0.231516734828E-0111 0. 297147086969E +00 27 -0.438977929243E +0412 -0.104964078480E +03 28 -0.315318636002E-0313 -0.775008265186E-02 29 0.139459067806E +0014 0.222564856042E +00 30 0.163298486259E-0615 -0.330783818273E+02 31 -0.326147254524E-0316 -0.313533565119E-02 32 0.342233333783E-01

26

DOE/CE/23810-33

Table 4. Coefficients to the MBWR equation of state for HCFC-123(units are K, bar, L, mol)

9 152 "_ 17p - ° .,-

n=l n=lO

Pc = 3.596417 mol/L

a1 = RTa2 = biT + b2T°.5 + b3 + b4/T + bs/T 2a3 = b6T + b7 + bs/T + b9/T2a4 = btoT + btt + b12/Ta5 = hi3

a6 = b14/T + bts/T 2a7 = b16/Tas = bl7/T + bla/T 2a9 = blg/T 2ato = b20/T2 + b21/T 3atl --- b22/T2 + b23/T4at2 = b24/T2 + b25/T3a_3 - b26/T2 + b27/'r4at4 = b28/T2 + b29/T3a15 = b30/T2 + b3l/T 3 + b32/T4

i bi

1 -0.193042434973E-01 17 0.106201732381E+002 -0.263410206086E+00 18 -0.401991529370E +023 0.266439262928E +02 19 0.156703568146E+0L4 -0.102447174272E +05 20 0.39581M226685E +075 -0.71496237_E +06 21 -0.490428403406E +096 0.179594735089E-01 22 0.171175389582E+067 -0.106601466621E +02 23 0.376067424212E + 108 -0.106973465680E +04 24 0.719667521763E+049 -0.150556666672E +07 25 -0.110348184730E+07

10 -0.12651M809410E-02 26 0.57121183795LE+0211 -0.123264787943E+00 27 0.642498617888E+0712 0.293238981229E +03 28 0.227383595657E +0113 0.134389339775E+00 29 -0.670239087161E+0314 0.745030119681E+01 30 -0.162446239669E-0L15 0.413916532768E+04 31 O.190850894641E +0216 -0.212267981526E +01 32 -0.267293932199E +04

27

DOE/CE/23810-33

Table 5. Coefficients to the MBWR equation of state for HCFC-124(units are K, bar, L, tool)

9 15

P = E anpn . exp(-P2/P_ ) E an02n'17n=! n_,10

Oc = 4.10153 mol/L

at = RTa2 = biT + b2T°-5 + b3 + b4/T + bs/T 2a3 = b6T + b7 + ba/T + b9/T2a4 = bloT + bl! + bt2/Ta5 = b13a6 = b14/T + bls/T 2a7 = bl6/Tas = b17/T + bls/T 2a9 = bt9/T 2alo = b20/T2 + b2t/T 3all = b22/T2 + b23/T4at2 = b24/T2 + b25/T3a!3 = b26/T2 + b27/T4at4 = b:8/T2 + b29/T3at5 = b3o/T2 + b31/T 3 + b32/T4

i bi

1 -0.204576807203E +00 17 -0.688566863825 E-012 0.183289763904E +02 18 -0.132391812938E+023 -0.436304129852E+03 19 0.667600131841E+004 0.784900629507E +05 20 -0.271799858829E +075 -0. 882621240790E +07 21 -0. I 11422740208E +096 -0.214052457908E-472 22 -0.175854504297E +067 -0.421490706906E +01 23 0.566801130630E + 108 O.379367628599E +04 24 -0.214018815397E +049 0.257319006570E +07 25 -0.327561948065E +0610 -0.128703560721E..02 26 -0.546930696467E +0211 0.318383860178E+01 27 0.931832376640E+0612 -0.126323679904E +04 28 0.19365497062 IE-0213 -.0.359253621024E-.01 29 -0.110844683745E+0314 -.0.201822160275E +02 30 -0.452370482664E-0215 0.239512195711E+03 31 0.163031126242E+0116 0.249923391219E+01 32 -0.681395650661E+03

28

DOE/CE/238i0-33

Table 6. Coefficients to the MBWR equation of state for HFC-125(units are K, bar, L, tool)

9 15

P--E anon* exp(-O2/0_)_ an02n-17n,,l n,,10

Pc = 4.7650 mol/L

aI = RT

a,2 = biT + b2T°.5+ b3 + b4/T+ bs/T2a3 = b6T + b7 + bs/T+ bg/T2a4 = bloT +btt + bt2/Ta5 - b13a6 = b14/T + bts/T 2a7 = bl6/Tas = b17/T+ b18/T2a9 = bt9/T 2ato = b20/T2 + b2t/T 3all = b22/T2 + b23/T4at2 = b24/T2 + b25/T3al3 - b26/T2 + b27/T'tal4 --" b28/T2+ b29/T3at5 = b30/T2 + b3t/T3 + b32/T4

i bi

1 0.695150135527E-01 17 -0.637258406198E-012 -0.109596263920E+02 18 0.291220108725E +023 0.2_Lg171467191E+03 19 -0.102197580663E+014 -0.510408655996E +05 20 -0.560938443772E +075 0.366753946576E +07 21 0.770104599552E +086 O.385350808228E-01 22 -0.224544749331E +067 -0.370988373715E +02 23 0.183452398750E + 108 0.134556555861E +05 24 -0.292476384933 E +049 0.371143622964E +07 25 -0.388467529252E +0510 -0.123685768773E-02 26 -0.339743229627E +0211 O.130495983411E+01 27 -0.544169038319E +0612 -0.468463056623E +03 28 -0.168305711698E +0013 0.511361375061E-01 29 0.115387298598E+0214 -0.204695459886E +02 30 -0.734893856572 E-0315 -0.414622181605E +04, 31 -0.329200834300E +0016 0.219744136091E+01 32 -0.403885226023E +01

29

DOE/CE/23810-33

THEORETICAL EVALUATIONS OF R-22 ALTERNATIVE FLUIDS:

Objective:

To provide information regarding the coefficients of performance (COP), capacities,compressor discharge temperatures, compressor discharge pressures, and compressordischarge pressure ratios of nine alternative fluids relative to HCFC-22 and threealternative fluids relative to R-502.

Results:

The Building Environment Division of the National Institute of Standards and Technology(NIST) completed this research under contract with ARTI. Detailed results of this studyare reported in the final report, DOE/CE/23810-7, Theoretical Evaluations of R-22Alternative Fluids, January 1993, by Piotr A. Domanski, PhD and David A. Didion,PhD. This report is currently available from the ARTI Refrigerant Database (RDB#3305, 32 pages). The following refrigerants and refrigerant blends were evaluated:

Alternative Refrigerants/Blends (% Weight)

HCFC-22 Alternatives

HFC-32/HFC- 125 (60/40)HFC-32/HFC- 134a (25/75)HFC-32/HFC-134a (30/70)HFC-32/HFC- 125/HFC- 134a (10/70/20)HFC- 32/HFC- 125/HFC- 134a (30/10/60)HFC-32/HFC-227ea (35/65)HFC-32/HFC- 125/HFC- 134a/R-290 (20/55/20/5)HFC-134a

R-290 (Propane)

R-502 Alternatives

HFC-32/HFC- 125/HFC- 143a (10/45/45)HFC- 125/HFC- 143a/HFC- 134a (44/52/4)HFC-125/HFC-143a (45/55)

Results of the evaluations are presented in Figures 7 and 8.

30

.............................................. nl............................................... ql[lllNIll...................................

DOE/CE/23810-33

Figure 7. Relative COPs and Capacities of HCFC-22 Alternatives.

Theoretical COP and Cal:)acitiesRelative to R-22

Rift Igol'an!(G(Imlelll_n)

RelativQ COP L__J_Relative CapJcltyA R3|/tll

teo/4o) 1.§

a R3|It||/134eI|OO _- i_

(10/I_1/10/I)G R_l|lt|l/t341 1.4

(10/I01t0)

o ,a,/,,., 1.2 _,i m=|/Isll+=4m 1 :/

(301tOIlOI _,,F R3|11341

(_o,',m 0.8G R31/134e

q_leml) 0.0N Illlao

i .1+4+ 0.4

0,2

A B C DIE P i O H I

Relative COP 0.97 _0.93i 0.93 ' 0.84i 0.98 0.99 ! 0.99i 0.94 i 0.98

Relative Caoaoity' 1.55 ! 1.18 1.04 1.03 _0.98! 0.93 i 0.88i 0.831 0.59t

Figure 8. Relative COPs and Capacities or R.502 Alternatives.

Theoretical COP and CapacitiesRelative to R-502

i Relative COP _ Relative CII)aOity i

1.2.

Refrigerant 1 _-(Compoe!tlon)

A R32/12fllt43e 0.8 _-

(10/4SI46)B Rt261143e O.O _-

(46/6fll

C R1261143el134a 0.4 _"(4415214)

0.2, _

0R32/125/143a R126/143a R125/143a/1341

+

Relative COP 0.97 0.93 i 0.93t 0.92Relative Capaolty i .1.13 0.04

31

DOE/CE/23810-33

CHEMICAL AND THERMAL STABILITYOF REI_IGERANT-LUBRICANT MIXTURES WITH METALS

Objective:

To provide information on the stability of potential substitutes for CFC refrigerants andappropriate lubricants.

Results:

Spauschus Associates, Inc., has completed this research under contract with ARTI. Adetailed report of results is presented in the final report, DOE/CE/23810-5, Chemicaland Thermal Stability of Refrigerant-Lubricant Mixtures with Metals, 9 October 1992,by Dietrich F. Huttenlocher, PhD, (RDB #3608, 126 pages). Key results aresummarized below:

A.[.ternat_e Refrigerant-Lubricant Combinatiom

CFC- I 1 (baseline) with:

naphthenic mineral oil (ISO 32)naphthenic mineral oil (ISO 46)

CFC-12 (baseline) with:naphthenic mineral oil (ISO 32)alkylbenzene (ISO 32)

HCFC-22 with:

naphthenic mineral oil (ISO 32)HFC-32 with:

pentaerythritol ester mixed-acid (ISO 32)polypropylene glycol butyl monoether (ISO 32)

HCFC- 123 with:

naphthenic mineral oil (ISO 32)naphthenic mineral oil (ISO 46)

HCFC- 124 with:

alkylbenzene (ISO 32)HFC- 125 with:

pentaerythritol ester mixed-acid (ISO 32)polypropylene glycol butyl monoether (ISO 32)modified polyglycol (ISO 32)

HFC- 134 with:

pentaerythritol ester mixed-acid (ISO 32)

32

.................................................................................................................................................................................................................... _mlllf ..........['11]_11..... IiI].........................

DOE/CE/23810-33

_Alternative Refrigerant-Lubricant Combinations (Continued)

HFC- 134a with:

pentaerythritol ester mixed-acid (ISO 22)pentaerythritol ester branched-acid (ISO 32)pentaerythritol ester branched-acid (ISO 100)polypropylene glycol butyl monoether (ISO 32)polypropylene glycol diol (ISO 22)modified polyglycol (ISO 32)

HCFC- 142b with:

alkylbenzene (ISO 32)HFC-143a with:

pentaerythritol ester branched-acid (ISO 32)HFC-152a with:

alkylbenzene (ISO 32)

Based on the results of his research, Dr. Huttenlocher made the following conclusions:

• All HFCs tested, along with HCFC-22, were very stable and did not undergo anymeasurable chemical reactions or thermal decompositions at temperatures up to 200°C(392°F).

• HCFC-124 and HCFC-142b were less stable than the HFCs tested but more stable than

CFC-12 (a long time industry standard).

• While HCFC-123 was the least stable of the "new" refrigerants tested, it was still tentold more stable than CFC-I 1 (the refrigerant it is intended to replace in low pressurechiller applications).

• The pentaerythritol ester lubricants included in the project exhibited acid numberincreases after aging at 200°C (392°F). The high viscosity (ISO 100) pentaerythritolester exhibited additional evidence of molecular changes during aging at 200°C. Theformation of CO2 indicated decarboxylation of the high viscosity pentaerythritol esterlubrication at that temperature.

• All of the polyalkylene glycol lubricants had signs of molecular change after aging.

i

33

I

DOE/CE/23810-33

MISCIBILITY OF LUBRICANTS WITH REFRIGERANTS

Objective:

To provide information on the miscibility of both current and new lubricants withpotential substitutes for CFC refrigerants.

Results:

Iowa State University of Science and Technology is performing this research undercontract with ARTI. Phase 1 of the project, preliminary miscibility screening, has beencompleted. These studies examined mixtures at three refrigerant-lubricant concentrations(10, 50, and 95% refrigerant by weight) and a single viscosity for each lubricant.Miscibility studies were conducted over a temperature range of-50 to 90"C (-58 to194°F) for most mixtures and -50 to 60°C (-58 to 140°F) for high pressure refrigerantmixtures. A detailed report on the results of this research is presented in DOE reportnumber DOE/CE/23810-6, Miscibility of Lubricants with Refn'gerants (Phase 1), October1992, by Michael B. Pate, PhD, Steven C. Zoz, and Lyle J. Berkenbosch (RBD #3503,64 pages).

Iowa State University has completed Phase 2 of the project which encompassed detailedmiscibility plots with five additional refrigerant-lubricant concentrations (20, 35, 65, 80and 90% refrigerant by weight) and two viscosity grades for each lubricant. The finalreport, DOE/CE/23810-18, Miscibility of Lubricants with Refrigerants, January 1994,by Michael B. Pate, PhD, Steven C. Zoz, and Lyle J. Berkenbosch, contains detailedresults. Preliminary results are summarized in Table 8.

34

DOE/CE/23810-33

Table 8. Miscibility of Lubricants with Refrigerants

,,, i i i || ,,.,, i ,,,.,

Refriqeranti ii iii i IBIB IIIII I IIIB BI II

Lubricant R22 R32 ' R123_ R124 ! R125 _ R134 R134aL R142b R143ai R152aE

iii IIIIB IIBII cl i iii iiii ii ii iiiiiiiii j ii i iMineral Oil > 10 : '> 20C > -40C,SO 32 cSt < 36% I M or ' I I I < 50% ' I t

> 90% ,.< 23% , > 80% ., ,_ , ,t ,.

Mineral Oil > 0 > -40C!> 50C ' _ > -30Ci

ISO 68 cSt or i I or or i I _ I I < 21% I I< 36% < 47% < 22% ! ;> 89%

ky ....... .... _.............AI Ibenzene ' ' ! , ,ISO 32 cSt M I i M i M I ' I I M I > 50C

Alkylbenzene : ; > 50C:ISO 68 cSt M I M ' M I i I I i M i I or

i ' < 20%

'PolypropyleneGlycol.... ' < _0C > -20Ci< 60C........Butyl Monoether M < 53% M M or _ or , or M < 35% MISO 32 cSt _ _ < 65% < 88% < 81%

Polypropylene Glycol < 20C < 40C i < 50C . < 80CButyIMonoether M < 47% i or M or i M i< 64% M < 38% < 80%

ISO 58 cSt > 21% < 65% , i,,, > 90%i i i ii i ii .

Polypropylene GlycolOiol M M M M M M M M i< 34% M

ISO 32 cSt , , _ _ i• _ ,_

Polypropylene Glycol .< 40C < 60C , ; < 70CDiol M < 49% M M ' or M or M < 48% < 81%

ISO 100 cSt ' < 80% < 66% , , > 90%i i

Modified Polyg'lycol "> -20C; < 60C > -10CI'< 30C i> OC [> OC > -40CiISO 32cSt < 23% > 10C , > -40C;< 37% i> 10C i < 23% < 22% :, < 23% i I M

> 50% < 21% i > 81% _< 20% > 79% > 52% > 68% ,_',,...................... "50CPentaerythrithol Ester < 50C > .

mixedacid M > 10C ' M M M M : < 69% i M ! < 38% ; M

ISO 22 cSt ' < 35% : > 91% __.._.__4_,Pentaerythrithol Ester > -20C; ! ;mixed acid M : or M M M M i M ; M < 49% M

'lSO32cSt _ < 51% _ ' ' i'Pentaerythrithol Ester < 60C ' > - 10C;

,,

< 35% : M M or . M or M < 36% ' M'mixed acid M i , i !IISO t00cSt , , , ,< 66% ; '< 64% i iii i Illll I

ilia r '_#e e ythrith I Ester : _ , _ !branched acid M > -20 M M M M M M ! < 51% _ M

, i

ISO 32 cSt : _ , ' ; : _PentaerythritholEster < 40C < 60C i I ;< 90C

branched acid M < 51% ' M M or M ! or i M , < 34% orISO 100..c.St .... < 77% : < 79% } ' < 90%,, , i i , ,, ,,i i

I - Immiscible or miscible only in a small temperature-concentration region.M - Miscible at all test temperatures and concentrations.

< *" - Miscible at all test temperatures or refrigerant mass concentrations below temperature or concentration indicated.> *" - Miscible at all test temperatures or refrigerant mass concentrations above temperature or concentration indicated.

35

DOE/CE/23810-33

COMPATIBILITY OF REFRIGERANTS AND LUBRICANTSWITH MOTOR MATERIALS

Objective:

To provide information on the compatibility of motor materials with potential substitutesfor CFC refrigerants and with suitable lubricants.

Results:

The Trane Company has completed this research under contract with ARTI. Detailedresults are presented in the final report, DOE/CE/23810-13, Compatibility of Refrigerantsand Lubricants with Motor Materials, May 1993, by Robert Doerr, PhD, Stephen Kujakand Todd Waite (Vol I - RDB #3857, 166 pages; Vol II - RDB #3858, 270 pages; VolIII - RDB #3859, 370 pages).

ReSults from the project indicate that most materials used in current hermetic motors arecompatible with the test refrigerant-lubricant combinations.

The project examined the compatibility of twenty-four hermetic motor materials witheleven pure refrigerants and seventeen refrigerant-lubricant combinations. Motormaterials tested included three types of magnet wires, six wire varnishes, six sheetinsulations, three sleeving insulations, three tie tapes, two lead wire insulations and onetie cord. A number physical property measurements were performed on samples of eachtest material before and after its exposure to the refrigerants and refrigerant-lubricantmixtures.

Refrigerants

HCFC-22 @ 90°C (194°F) HFC-134 @ 90°C (194°F)HCFC-123 @ 90°C (194°F) HFC-32 @ 60°C (140°F)HCFC-124 @ 90°C (194°F) HFC-125 @ 60°C (140°F)HCFC-142b @ 90°C (194°F) HFC-143a @ 60°C (140°F)HFC-152a @ 90°C (194°F) HFC-245ca @ 121°C (250°F)HFC-134a @ 90°C (194°F)

36

........................... ..... : ......................................... | ........................................ IllrIF"_--"

DOE/CE/23810-33

Refrigerant-Lubricant Combinations at 127°C (260°F)

HCFC-22/mineral oil (ISO 32)HFC-32/polypropylene glycol butyl monoether (ISO 32)HFC-32/pentaerythritol ester branched-acid (ISO 32)HCFC-124/alkylbet_ene (ISO 32)

" HFC-125/polypropylene glycol butyl monoether (ISO 32)HFC-125/modified polyalkylene glycol (ISO 32)HFC-125/pentaerythr_tol ester branched-acid (ISO 32)HFC-134/pentaerythritol ester branched-acid (ISO 32)HFC-134a/polypropylene glycol butyl monoether (ISO 32)HFC-134a/polypropylene glycol diol (ISO 32)HFC-134a/modified polyalkylene glycol (ISO 32)HFC-134a/pentaerythritol ester mixed-acid (ISO 22)HFC-134a/pentaerythritol ester branched-acid (ISO 32)HCFC-142b/alkylbenzene (ISO 32)HFC-143a/pentaerythritol ester branched-acid (ISO 32)HFC-245ea/pentaerythritol ester branched-acid (ISO 32)HFC- 152a/alky lbenzene

Motor Materials Evaluations_

Varnish Spiral Wrapped Sleevingweight change weight change

break loan strengthLead Wire

weight change Sheet Insulationdielectric strength weight change

tensile strengthTie Cord elongationweight change dielectric strengthbreak load strength

TapesMagnet Wire/Varnish weight change

- bond strengthburnout resistance

dielectric strength

There were no compatibility concerns with any of the three magnet wires tested. Mostof the test varnishes were compatible with the refrigerant-lubricant mixtures. One of thesix tested varnishes, the Sterling Y-833 varnish (100% solids VPI epoxy), raised

m

37

DOE/CE/23810-33

compatibility concerns. It was considered incompatible with HCFC-123 and exhibitedphysical changes when tested with HCFC-22. The varnish became soft, limp and crazedafter the 500-hour exposure to HCFC-123. The varnish also became severely crazed andlimp after exposure to HCFC-22. Varnish is used in hermetic motors to bind motor wirewindings and to prevent wire-to-wire rubbing from stripping away the insulating coat andelectrically shorting the motor.

Only one of the three tapes tested displayed any compatibility problems. Theglass/acrylic tape was considered incompatible with HCFC-123. After exposure, itexhibited a large weight loss, turned green in color, rolled up and separated from itsbacking. Compatibility concerns also arose in tests with nine of the seventeenrefrigerant-lubricant mixtures. After exposure, the tape curled up and its backing easilyrubbed off. However, when the tape was heated for an addition 24 hours at 150°C(302°F) it regained its original unexposed form.

Three of the six sleeving materials tested had compatibility concerns. The laminatingadhesive in the Nomex, Mylar, and Nomex/Mylar sleeving insulations weakened afterexposure to HCFC-22/mineral oil and/or HCFC-124/alkylbenzene mixtures. However,it was noted that these materials have been used in HCFC-22/mineral oil applications for20 to 30 years without equipment reliability problems.

Sheet insulation materials raised more compatibility concerns than any of the other=

materials tested. The Nomex/Mylar/Nomex was considered incompatible with the HFC-134a/polypropylene glycol diol (PAG-diol) mixture. The adhesive which bonds thelayers together dissolved. Pockets of delamination also resulted after the material wasexposed to five of the pure refrigerants and eleven of the refrigerant-lubricant mixtures.The material also lost flexibility or became brittle after exposure to four otherrefrigerant-lubricant mixtures.

Dacron/Mylar/Dacron sheet insulation was also considered incompatible with the HFC-134a/PAG-diol mixture because of dissolution of the laminating adhesive. Additional

compatibility concerns were raised due to excessive weight loss after exposure of thematerial to HCFC-22, HFC-245ca, HFC-134a/polypropylene glycol (PAG-butylmonoether) and HFC-134a/modified PAG mixtures. The material also experiencedembrittlement and/or lost flexibility after exposure to four other refrigerant-lubricantmixtures.

Likewise, Melinex 228 and Mylar MO raised compatibility concerns due to embrittlementor loss of flexibility after exposure to four refrigerant-lubricant mixtures which containedmineral oil or alkylbenzene. Nomex 410 and Nomex 418 raised compatibility concernsbecause of excessive weight loss after exposure to HFC-125.

38

DOE/CE/23810-33

COMPATIBILITY OF REFRIGERANTS AND LUBRICANTSWITH ELASTOMERS

Objectives:

• To provide compatibility information for elastomers with potential substitutes for CFCrefrigerants and with suitable lubricants.

• To obtain data on changes in the physical and mechanical properties of selectedelastomers after thermal aging in refrigerant-lubricant mixtures.

Results:

The University of Akron has completed this research under contract with ARTI.Detailed results are presented in the final report, DOE/CE/23810-14, Compatibility ofRefrigerants and Lubricants with Elastomers, January 1994, Gary R. Hamed, PhD,Robert H. Seiple, and Orawan Taikum.

This research project examined the compatibility of ten refrigerant and seven lubricantswith ninety-five elastomeric materials:

Refrigerants Lubricants

HCFC-22 naphthenic mineral oil (ISO 32)HCFC-123 alkylbenzene (ISO 32)HCFC-124 polypropylene glycol butyl monoether (ISO 32)HCFC-142b polypropylene glycol diol (ISO 32)HFC-32 modified polyglycol (ISO 32)HFC-125 pentaerythritol ester, mixed-acid (ISO 22)HFC-134 pentaerythritol ester, branched-acid (ISO 32)HFC-134aHFC-143aHFC-152a

39

DOE/CE/23810-33

Elastomer Families

butyl polypropylene TPE (1 type) nitrile rubbers (10 types)butyl rubbers (7 types) polychloroprenes (2 types)chlorinated polyethylenes (3 types) polyisoprenes (3 types)chlorosulfonated polyethylenes (5 types) polysulfide rubbers (4 types)epichlorohydrin based rubbers (6 types) polyurethanes (7 types)ethylene acrylic elastomers (2 types) silicones (5 types)ethylene propylene rubbers (3 types) styrene butadiene rubbers (4 types)ethylene propylene diene rubbers (5 types) thermoplastic elastomers, TPEsfluorinated rubbers (7 types) (11 types)

plus, ten industry-supplied gaskets of various compositions

Swell behavior ofelastomer samples were determined by comparing pre-exposure samplemeasurements for weight, thickness and diameter with their measurements after exposure.As indicated above, these elastomeric formulations included general purpose and specialtythermoset and thermoplastic elastomers.

Refrigerant Immersion Studies: Elastomer samples were completely immersed in the testrefrigerant, sealed in a pressure vessel and maintained at room temperature (ambient) for14 days. In situ diameter changes were determined using a traveling microscope after24-hour, 72-hour and 14-day exposures. Following the 14 day exposures, the sampleswere remeasured 2 hours and 24 hours after they were removed from the pressurevessels.

In reviewing the results, the following general statements can be made concerning in situswelling measurements after the 14 day exposures:

• samples exposed to HCFC-123 had the largest swell,

• samples exposed to HCFC-22, HCFC-124, HCFC-142b had moderate swell,

• samples exposed to HFC-32, HFC-125, HFC-134, HFC-134a, HFC-143a, andHFC-152a had the least swell.

Refer to Table 9 for a relative comparison of in situ swelling results.

40

DOE/CE/23810-33

Lubricant Immersion Studies: Elastomer samples were completely immersed in the testlubricant, sealed in a glass vessel and then heated at 60°C (140 °F) for 14 days. Samplediameters were measured in situ after 24 hours of exposure. The elastomer samples werealso measured for weight, thickness and diameter immediately after the 14-day exposureand then again 24 hours after removal.

Several of the elastomeric compositions, including some of the industry-supplied gaskets,were resistant to swelling in all of the lubricants. These included rubbers from theepichlorohydrin, nitrile, polysulfide rubber, and thermoplastic elastomer families. Referto Table 6 for a relative comparison of the in situ swelling results.

Refrigerant-Lubricant Thermal Aging Tests: Based on the results of the separatelubricant and refrigerant studies, twenty-five elastomeric samples were selected forinclusion in refrigerant-lubricant thermal aging tests. These elastomers were individuallyimmersed in seventeen separate refrigerant-lubricant mixtures for 14 days at 100 °C (212°F). Depending on the refrigerant-lubricant combination, the refrigerant weight percentvaried from -20%.to 50% concentration to maintain a vapor pressure of 275-300 psia.After the 14-day exposures, dimensional, hardness, and tensile values of the exposedelastomers were obtained and compared to those of non-aged specimens.

As a general trend, it was found that the tensile strengths of the aged elastomers wereinversely related to the amount of swelling they exhibited after aging in the refrigerant-lubricant mixtures. When swelling was large, elastomer tensile strength decreaseddramatically. However, in some cases, when swelling was slight or negative (i.e.,shrinkage from material extraction) tensile strength increased after aging. In all cases,filled rubbers showed less change of tensile strength after aging compared to unfilledcounterparts.

41

DOE/CE/23810-33

Table 9. Relative in situ Elastomer Swelling

RELATIVE IN SITU ELASTOMER SWELLING: REFRIGERANTS

22 32 123 124 125 134 134s 142b 143a 152==,,

butyl polypropylene TPE - s s - s s s - s -butyl rubbers S s - s S s s s s schlorinated polyethylenes S S - s S s s s S schlorosulfonated polyethylenes - s - s s s s s s sepichlorohyddn baaed rubbers L S L L S - S S S SEPM rubbers s s - s s s s s s s

ethylene ecrlic elestomers L - L L L - - L S -ethylene propylene diene rubbers S S - s S s s s _ sfluorinated rubbers L - L L - L L L - L

nitrile rubber= L S L L S - S - S -

polychloropr=nes S s - s s s s s s spolyisoprenes - s L S S S S - S Spolysulflde rubbers S s L S S S S S S Spolyurethanes L s L L - - s - s -silicones L - L L - - - L - L

styrene butadiene rubbers - s L S S S S S S Sthermoplutic elaatomers (TPE) - s - - s s s s s s

RELATIVE IN SITU ELASTOMER SWELLING" LUBRICANTS

AB MO PEBA PEMA PPGBM PPGD MPG

butyl polypropylene TPE - - s s s s sbutyl rubbers L L S S S S Schlorinated polyethylenes s - - - s s schlorosulfonated polyethylenes .... s s sepichlorohyddn baaed rubbers s s - - s - sEPM rubbers L L S S S S S

ethylene scrli¢ elestomer= - - L L L L L

ethylene propylene dlene rubbers L L S S S S Sfluorinated rubbers s s L - S S Snitrile rubber= s s - - s s spolychloropr=nes - - - L - S Spolyisoprenes L L - - s s Spolysulfide rubbem s s s s s s spolyurethanes - s s - s - -silicones - - - s s s sstyrene butadiene rubberI L L - - - S Sthermoplastic elaatomer= ('I'PE) s - s s s s s

!ecjencl:S - smaJl linear _welis; less _arl 8 %

L - large linear swells; greater than than 35 %- - mixesswellvalue4Jsneer 8% < swell< 35%

AB - ali(yibenzeneMO - mineralodPEBA - Pentaery_rrtolesterorinchea ¢lcIcIPEMA - penlery_ntol emmrmrxeaac=clPPGBM- Dolypropyleneglycolioutylmonoett_rPPGD - i:x_yl:x'ooy_megly¢o(MPG - mOOWfleClpoty_y¢O_ j

42

DOE/CE/23810-33

COMPATIBILITY OF REFRIGERANTS AND LUBRICANTSWITH ENGINEERING PLASTICS

Objectives:

• To provide compatibility information for engineering plastics with potential substitutesfor CFC refrigerants and with suitable lubricants.

• To obtain data on changes in the mechanical properties of selected plastics after thermalaging in refrigerant-lubricant mixtures.

Results:

Imagination Resources, Inc., has completed this research under contract with ARTI.Detailed results are presented in the final report, DOE/CE/23810-15, Compatibility ofRefrigerants and Lubricants with Engineering Plastics, December 1993, by Richard C.Cavestri, PhD.

This research project examined the compatibility of ten refrigerants and seven lubricantswith twenty-three engineering plastics:

Refrigerants Lubricants

HCFC-22 naphthenic mineral oil (ISO 32)HCFC- 123 alkylbenzene (ISO 32)HCFC-124 polypropylene glycol butyl monoether (ISO 32)HCFC-142b polypropylene glycol diol (ISO 32)HFC-32 modified polyglycol (ISO 32)HFC-125 pentaerythritol ester, mixed-acid (ISO 32)HFC-134 pentaerythritol ester, branched-acid (ISO 22)HFC-134aHFC-143aHFC-152a

43

DOE/CE/23810-33

Engineering Plastics Tested.

acetal polybutylene terephthalate (PBT)acrylonitrile-butadiene-styrene(ABS) polycarbonateliquid crystal polymer (LCP) polyetherimidemodified polyetherimide polyethylene terephthalate (PET)modified polyphenylene oxide polyimide thermoset (2 types)nylon 6/6 polyphenylene sulfide (PPS)phenolic polyphthalamidepolyamide-imide (2 types) polypropylenepolyaryletheretherketone (PEEK) polytetrafluoroethylene (PTFE)polyaryletherketone (PEK) polyvinylidene fluoridepolyarylsulfone

Lubricant Immersion Studies: The plastic specimens were evaluated after 14-dayexposures in pure lubricants at 60°C (140°F) and 100°C (212°F). Each plastic wasaffected to some extent by the lubricants. In general, weight and dimensional changeswere in the plus or minus 1-2% range. However, the ABS specimens exhibited relativelylarger changes in all the lubricants (in the 5-15 % range).

Refrigerant Immersion Studies: The plastics were evaluated at ambient room temperatureand 60°C (140°F) in pure refrigerant for 14 days at the saturation pressure of therefrigerant. All refrigerants had some effect on the plastics; generally, weight increaseand some softening of the plastics. HFC refrigerants seem to have the least effect on theplastics. The ABS plastic failed (e.g., dissolved or deformed) in HCFC-22, HFC-32,HCFC-123, HCFC-124, HFC-134, and HFC-152a. The polycarbonate and the modifiedpolyphenylene oxide plastics failed in HCFC-123.

Stress Crack-Creep Rupture Tests: Linear creep was measured for plastic test barssubmerged in an ISO 32 cSt branched acid polyolester lubricant with 40% refrigerantconcentrations (by weight) at 20°C (68°F) for 14 days. Each plastic was weight loadedat 25 % of its ultimate tensile capability to stress the gage area of specimen test bars.The resultant deformation under load information provided the creep modulus arisingfrom the exposure effects of synthetic lubricants with the differing refrigerants.

Plastic creep appeared to be nearly the same for all refrigerants. However, plasticsexposed to HCFC-22 exhibited slightly lower creep rates than when exposed to the othernine refrigerants. Two plastics that routinely failed (e.g., broke within one hour) wereABS and modified polyphenylene oxide. HCFC-123, as expected, induced a pronouncedincrease in plastic creep, but did not promote rupture of the r!astic test specimens.

44

1

DOE/CE/23810-33

Refrigerant-Lubricant Therma_l Aging Tests: Thermal aging tests on the twenty-threeplastic specimens in seventeen refrigerant-lubricant combinations were completed. Thesetests were performed for 14 days at 150°C (300°F) and at refrigerant pressures from1,900 to 2,070 kPa (275 to 300 psia). Due to its higher reactivity, HCFC-123 agingtests were performed at 125°C (260°F) and at 105°C (220°F). Physical changes wereobserved, dimensional changes measured, and specimen tensile properties were comparedto the original, unexposed specimens.

After aging, the plastics exhibited minimal dimensional and weight changes (i.e.,generally within plus or minus 2%). However, the phenolic, polyvinylidene fluoride,and polypropylene plastic specimens exhibited the largest dimensional and weight changes(generally 5-20%). As compared to the tensile tests performed on non-aged plastic testbars, the aged specimens exhibited large reductions in tensile capabilities (i.e., changesin tensile strengths ranged from a 30% gain to a 50% loss, changes in elongation rangedfrom a 10% increase to a 85% loss). Hence, as a result of environmental embrittlement,many plastics broke after a much smaller elongation under a much lower tensile load; ascompared to the non-aged specimens.

45

DOE/CE/23810-33

ELECTROHYDRODYNAMIC (EHD) ENHANCEMENTOF POOL AND IN-TUBE BOILINGOF ALTERNATIVE REFRIGERANTS

Objectives:

• To construct a test rig that can measure improvements with in-tube boiling and in-tubecondensation heat transfer performance when utilizing EHD enhancement technology.

• To ascertain the heat transfer b,enefits on pool boiling with HCFC-123/lubricant on singleand multiple enhanced tubes when utilizing EHD techniques.

Results:

The University of Maryland completed this research under contract with ARTI. Thefinal report detailing the pool boiling test results and the fabrication and qualification ofthe in-tube apparatus is available under DOE report number DOE/CE/23810-17, EHDEnhancement of Pool and In-Tube Boiling of Alternative Refrigerants, August 1993, byM. M. Ohadi, S. Dessiatoun, A. Singh, and M. A. Faani (RDB #3A16, 62 pages).

This project accomplished three major tasks: (1) literature search on prior EHD research,(2) EHD pool boiling experiments with HCFC-123 and HFC-134a, and (3) design,fabrication, and shakedown of an EHD in-tube boiling/condensation test rig.

For pool boiling, higher applied electric potentials resulted in higher EHD-inducedeffects that promoted refrigerant bubble break-up and increased bubble departure speeds;collectively leading to higher heat transfer rates. For pool-boiling with HCFC-123 andHFC-134a, it was reported that the heat transfer rates increased 5 - 8 fold, as comparedto the non-EHD enhanced runs. This depended on whether or not 2% lubricantconcentration was added and on whether mesh-type or straight-wire electrodes wereutilized.

46

[ I

DOE/CE/23810-33

CO_LIANCE,WITH AG_EME_

ARTI has complied with all terms of the grant agreement during the reported period.

PRINCIPAL INVESTIGATOR'S,EFFORT

Mr. Mark Menzer is the ARTI principal investigator for the MCLR program. During the fourthquarter of calendar year 1993, Mr. Menzer devoted a total of 192 hours (44.2% of his availablework hours) on the MCLR program.

47

j II

t

.

I

II

_ ii I